Moonshot AI Seeks Hong Kong IPO as Private Round Could Value It Above $30B

Moonshot AI, the Beijing startup behind the Kimi chatbot, has asked its own investors to clear the way for a Hong Kong initial public offering that could arrive within six months. The company circulated a shareholder resolution seeking that approval, The Information reported, even as it finishes a private funding round that people close to the talks say could value it at more than $30 billion. If the timeline holds, Moonshot would become the third of China’s frontier model labs to put a valuation in front of public-market investors, following Z.ai and MiniMax onto the Hong Kong exchange within roughly a year of them.

The request lands days after Moonshot released Kimi K3 on July 16, 2026, an open-weight model with 2.8 trillion parameters and a one-million-token context window that debuted at No. 3 on the Artificial Analysis leaderboard and knocked several US chip and AI stocks lower in the days that followed. A benchmark-driven news cycle is exactly the moment a private company wants to reprice itself, and Moonshot is moving to convert the attention into a listing before it fades.

Inside the listing plan

The shareholder resolution is the procedural first step: existing backers have to bless the plan before bankers and regulators get involved. Bloomberg reported that Moonshot told investors it is preparing to list in as little as six months, and that its annual recurring revenue reached $300 million in June 2026, up from $200 million two months earlier.

To qualify for a Hong Kong debut, Moonshot is unwinding the offshore holding structure that Chinese startups use to take in foreign capital, a setup it signaled it would dismantle in May to satisfy Beijing’s tightened rules on overseas listings. A mainland-friendly joint-venture arrangement replaces it while preserving foreign investors’ stakes. Hong Kong has become the default venue for Chinese AI floats as US markets stay effectively closed to mainland tech names, and the city’s listing regime lets pre-profit technology companies apply once they clear a modest revenue bar that Moonshot has already passed.

What the numbers do and don’t show

The growth figures are the argument Moonshot is making to shareholders, and they are steep: a 50% jump in recurring revenue over two months, from paid Kimi subscriptions and enterprise API usage. But annual recurring revenue is a run-rate the company reports itself, not an audited result. It says nothing about gross margins, the cost of the GPUs serving Kimi, customer retention, or how much cash the company burns training each new model. A Hong Kong prospectus would force all of that into the open for the first time, which is precisely what public-market investors will price on once the marketing cycle cools.

The valuation trajectory is just as fast. Moonshot was worth roughly $4 billion at the end of 2025. In May 2026 it raised about $2 billion at a valuation near $20 billion, a round led by Meituan’s investment arm. The $30 billion figure now attached to the closing round is not confirmed, and the round has not closed. Chief executive and co-founder Yang Zhilin, a former Tsinghua professor who worked at Meta and Google, has told investors the company holds more than 10 billion yuan, about $1.4 billion, in cash and is under no pressure to raise. That framing matters: it positions the IPO as opportunistic timing rather than a scramble for liquidity.

The race to set China’s AI price

Moonshot is not listing into an empty field. It is listing into one with prices already on the board: Z.ai and MiniMax both trade in Hong Kong, and DeepSeek, its closest rival in cheap open-weight models, is weighing an IPO of its own in 2027, while Alibaba took its flagship Qwen model open-weight the same weekend Moonshot’s IPO plans surfaced. What Moonshot is racing for is not the first listing but the price-setting one: at a $30 billion ask it would become the number public markets use to value every Chinese lab built on open-weight releases, and the earlier, smaller debuts would be repriced against it.

Its cap table is its strongest signal. Alibaba and Tencent, rivals across cloud, commerce, and payments, both sit on Moonshot’s register alongside HSG, IDG Capital, and 5Y Capital. Two companies that rarely agree backing the same lab is the kind of convergence that gives a listing story credibility with outside investors.

The open questions are the ones a filing would answer. Whether the funding round closes at $30 billion, whether Moonshot files inside its six-month window, and whether the K3-driven enthusiasm survives contact with audited numbers will decide if this becomes the reference IPO for Chinese AI or an example of a valuation set at the top of a hype cycle. For now, the company has the growth, the cash, and the attention. What it does not yet have is a public price.